[N4NV]

I think my turret woes may have something to do with the thick oil caused by cold temperatures that cause the turret to turn a little slower and just missing the index. I ordered new O-rings from Hardinge anyway. Their price for O-rings it good, but there is a minimum $30 order. I ordered every non-standard O-ring I could find in the catalog to bump the order to over $30.

While I wait for the parts, I am moving forward with other little items that need to be addressed. Neither of the electrical enclosures on the ends of the stand have any kind of air venting. Now that I have two VFDs in one box and my computer in the other, I new I had to address that issue.

I dropped by my local surplus store (Halted) and picked up a couple of muffin fans and grills. I installed one fan and one inlet grill in each enclosure. I have the fans pulling air out of the enclosure and mounted the inlet grills (with foam filters) on the outside and opposite the fans. Having the inlet grills on the outside (and the front of the machine), lets me see more readily if the grills need cleaning.

Vince

[N4NV]

I did not like the way my home made way wipers worked out so I ordered the proper ones from the factory. It was about $40 for the front wiper and $50 for the rear. They arrived a couple of days ago so I installed them. They do a much better job than my home made ones. Now I am not so worried about the Teflon getting scored under the carriage.

While I wait for the O-rings to arrive for the turret cushion, I am working on some Hall effect limit switched for my Z axis on my big mill. I have some mechanical limit switches now, but at certain RPMs they trip, causing Mach to do an e-stop which then causes a lost position. These Hall switches (the same ones I used for the turret encoder) should be impervious to vibration.

Vince

[N4NV]

Unfortunately, all work on the lathe has stopped for now. I got new O-rings from Hardinge for the turret and installed them, but suddenly, Mach stopped communicating with the Cubloc PLC. At the same time I lost control of the spindle speed. It now will only give me about 30% of commanded speed and the speed changes when I move the mouse. I purchased a new Cubloc board thinking that might be the problem, but is was not. I now have purchased a new computer to see if that was the problem. I will not get the computer for at least another week so everything is on hold until then.

Vince

[N4NV]

Oh well, in the words of Emily Latilla, never mind. Hood suggested I check the pulse length again so I did. I changed it to 4 (it was at 2) and now the spindle speed started working again. I don't know how it got changed to 2 or if it was ever at 4, at least now it works.

That got me thinking about the Cubloc. I purchased a new serial cable and installed it and now I can run the turret again! I guess I pinched the serial cable in the cabinet door.

I still can't index only one tool position. I am pretty sure it has something to do with the temperature and thickness of the lubrication. For now, I will have to make sure I change at least two tool positions or make two tool changes to get to a tool.

Too bad my 10 day vacation ends today. I was going to install my spindle sensor and work on threading, maybe install my smoothstepper. Now they will all have to wait. I also fixed my limit switch tripping problem on my mill today. I installed a capacitor across the inputs and now I don't get anymore trips (at least in the 3 test I ran).

Vince

[N4NV]

My spindle speed control issues still exist. It is either the computer or the break out board. Regardless of the settings, sometimes it works and sometimes it does not. While I wait for the new computer to arrive, I made up my spindle index sensor.

If you have not followed my other thread about my big milling machine, then this will be new to you. I made an index sensor out of a hall switch and magnet. Using Delrin, I milled a mount to hold the magnet. After installing the magnet mount on the spindle shaft, I made a Delrin mount for the Hall sensor. I milled some grooves in the Delrin for the wires to fit. I used epoxy to initially hold the wires in place, then I covered them and the sensor with epoxy putty. After curing a light sanding cleaned things up a bit.

I had a spindle encoder shaft cover from that other HNC I have been gutting. I cleaned it up and painted it. It worked well for a place to mount the Hall sensor. I still have to machine a cover for it, but until I get the spindle speed control working again, there is no hurry.

I tested the sensor and it worked well. If you ever install an index pulse sensor with Mach, remember to set the index pulse debounce to 0 or it will not work correctly. I found out the had way.

Vince

[Runner4404spd]

VInce, on your spindle issues my mill seemed to do the same thing for a while when i was playing with the Mach setup. i found the Step pulse on the motor tuning page would be all over the place. i finally had to set it 5 for it to work well the cnc4pc speed control board and the vfd. just a though you may want to check it out.

[N4NV]

VInce, on your spindle issues my mill seemed to do the same thing for a while when i was playing with the Mach setup. i found the Step pulse on the motor tuning page would be all over the place. i finally had to set it 5 for it to work well the cnc4pc speed control board and the vfd. just a though you may want to check it out.

My step pulse has been set at 5 for weeks now. Still very erratic behavior.

Vince

[BillTodd]

Hi Vince,

That looks like unscreened Cat5 cable that you've used for the sensor. It might be worth replacing it with a good quality screened cable (from the sensor box to the PC). If the BoB or PC input is edge sensitive it could be triggering on noise picked up by the cable.

Another option is to filter the power to the hall effect device with a small capacitor at the sensor end.

[N4NV]

Hi Vince,

That looks like unscreened Cat5 cable that you've used for the sensor. It might be worth replacing it with a good quality screened cable (from the sensor box to the PC). If the BoB or PC input is edge sensitive it could be triggering on noise picked up by the cable.

Another option is to filter the power to the hall effect device with a small capacitor at the sensor end.

The Hall effect device works perfectly, it has nothing to do with the speed control to the VFD. In two years I have never had a problem with any device I have connected with cat 5 cable (mostly encoders). The speed control comes right off the BOB with 0 to 10 V to the VFD. There is only 1 cable from the speed control to the VFD and it is 4 conductor shielded. I measure only 2.8V from the BOB, it should be 10V.

Vince

[BillTodd]

Sorry I saw pics of the spindle speed encoder and jumped to the wrong conclusion.

[N4NV]

Sorry I saw pics of the spindle speed encoder and jumped to the wrong conclusion.

That's OK, at this point I will take any suggestions, even something obvious that I might have missed. I am leaving on Thursday for the Cabin Fever expo and Mach seminar. I won't get a chance to work on the lathe for two weeks. my replacement computer showed up today. Oh well, it should be fun at the expo (except for the cold, this transplanted Californian no longer likes it much).

Vince

[N4NV]

I went to the Cabin Fever model show and Mach seminar so I have not been able to work on the lathe for a few weeks. It looks like I will be laid off in the next couple of days so I will have a lot of time to work on the lathe and get my garage cleaned up.

I made a cover for the spindle sensor housing. It took over 2 hours because I could not believe the screw location measurements I was getting. I used the video camera plug in with Mach to locate the holes. I can measure to +-.001 inch with my camera. Only a couple of the holes came out to an even fractional measurement. After many measurements I went ahead and drilled it with the measurements I made and it came out perfect.

I decided that I am having to many problems with my break out board (I just found a dead input), so I am removing it and installing a PMDX-122 along with a smooth stepper and a Peter Homann spindle speed control. This will be a major change as I have never used a smooth stepper board.

Vince

[N4NV]

Over the Christmas holiday, I spent some time scanning the Hardinge HNC maintenance manual and also the CHNC parts manual (neither were copy written). Both of the documents apply to the lathe that I have been working on. There is only a little difference between the HNC and CHNC maintenance manual, mostly in the stand. The documents are in PDF form and have been through OCR so they are text searchable. I am selling the manuals on CD for $20. If you would like one send me a PM. I will also include the Hardinge tool catalog that applies to the CHNC.

Vince

[N4NV]

If you have read the thread, you know that I have been having problems with the speed control of my spindle/VFD. I am using a CNC4PC C11G board that has an integral speed control on it. I also found that I had a bad input on the board as well, pin 11. Arturo said the problem has to be a problem with my computer since my mouse was affection the speed of the spindle. I purchased a new computer, but after talking to several other, I decided to try a new break out board as well. I purchased a PMDX-122 BOB as well as a speed control board (DigiSpeed-SD) from Homann Designs.

I did not install the new computer but installed the PMDX and DigiSpeed boards today. I now have my spindle speed back. The only thing I have to get straight on the DigiSpeed is the CW and CCW relays. When making adjustments to Mach, I found that by varying the "steps per" motor tuning section, I could adjust the linearity of the speed over the full range. I started with 1000 and found that a setting of 785 gave me a speed within 1% of what was commanded.

As an added bonus to installing the PMDX board, I got much faster rapids as well. Something on the CNC4PC board was causing noise on the motor outputs which limited my rapids to 110 IPM. With the PMDX board I can now rapid at 150 IPM! That's only a couple of seconds for full travel of the Z axis. I turned it down to 130 IPM for some safety margin. I also set the X axis for 90 IPM since the travel is usually only a couple of inches.

In addition to the PMDX board and the DigiSpeed board I also picked up a smooth stepper board. I installed it next to the other boards but have not hooked it up yet. I want to make sure I have everything working before I add another variable to the mix.

Things are looking up again.

Vince

[N4NV]

I got the CW and CCW relays working on the DigiSpeed controller. I should now be able to tap with a floating tap holder (I picked up a new Hardinge floating tap holder for $150 on eBay). At under 500 rpm I can change direction in under 1 revolution of the spindle. If I am running CW with an M3 command I can directly enter an M4 command without a hiccup from the PLC.

With the PMDX BOB up and running I have an extra input available. I hooked up my original turret hall sensor (the one that detects if the turret seats all the way) to the BOB. With the PMDX board no pull up resistors are required to get the Hardinge Hall switches to work properly.

With the turret sensor hooked up, I elisted some help from Jason and Scott over at the Mach support forum. Jason wrote me a script for a tool change macro and Scott help with a brain (a brain is a ladder like program used in Mach). With the new script and brain, when I enter an M6 command, the saddle will move to the tool change position (G53 Z-0.2) and then the PLC will perform the tool change. If the turret does not seat all the way (like when I change only one tool position) it will not proceed with the Gcode program and ask that the operator seat the turret. Once seated the operator can click OK and Mach will proceed with the rest of the Gcode program. This was a big safety item I wanted to get done. I hit the chuck twice with a tool when I forgot to move to the tool change position before doing a tool change.

I installed a new computer, another SX270 because Mach was locking up occasionally. I was told when you install Windows XP not to install any of the service packs. I have been updating XP with the service packs and I think some of my problems started after one of my updates. I am now going to re-format the drive on the old SX270 and re-install XP. I will then put that computer in my mill and take the computer in my mill, an SX260, and put it in my Bridgeport, which has an older GX260 with a dead internal battery.

Vince

[phomann]

The only thing I have to get straight on the DigiSpeed is the CW and CCW relays. When making adjustments to Mach, I found that by varying the "steps per" motor tuning section, I could adjust the linearity of the speed over the full range. I started with 1000 and found that a setting of 785 gave me a speed within 1% of what was commanded.

Vince

Hi Vince,

Good to see things are working out. In producing an analog voltage out of the DC-06 DigiSpeed-SD the step/inch in the spindle motor setup tab work like this.

The DC-06 passes the step pulse through a monostable. This converts the narrow step pulse into one of a fixed and known width. The advantage of this is that when you change the step pulse width in Mach, the analogue voltage is not affected.

The step pulse generated from the monostable then feeds a RC charge pump. Think of the step pulse as a small charge of energy. The more pulses you generate the more energy goes into the charge pump and the higher the output voltage that is generated.

In the end you can only produce a voltage that is slightly less that the supply being fed to the DC-06 from the VFD (say 10V). If you set the steps/inch too high then the output voltage hits the 10V ceiling too early. As a higher step pulse stream is supplied, the output voltage is clamped to the 10V thus creating the non-linearity.

What you want is for the maximum steps/inch setting to supply just enough energy to the charge pump to reach the 10 volts. If it is too much the spindle will run faster that the set speed. If there are not enough pulses, the spindle will not reach the set speed.

What you need to do is what you have done. Play around with the steps/inch setting until the maximum actual spindle speed is reached at the same time the maximum set spindle speed is reached.

I hope this helps explain what is going on.

Cheers,

Peter.

[N4NV]

Thanks for the explanation Peter. I figured out what is was doing by playing around with the settings, I did not know what was going on inside the "box". I ordered another one for my mill.

Vince

[N4NV]

I can't keep my collet closer valve from leaking air. If anyone has a suggestion for a replacement valve, I would appreciate it.

Thanks

Vince

[Karl_T]

Vince, are you referring to the solenoid valve in the lower cabinet left side? You don't want to ask Hardinge, HUGE $$$. Mine leaked also. A complete disassembly.cleaning, and replacement of O rings fixed it.

Karl

[Runner4404spd]

i believe the valve in question is a mac valve. the valves cost about $45.00 direct from mac. if its not a mac i recommend getting one from them to replace it. hardinge went to mac valves on all my machines. i have a chnc and cobra. the solenoid should have a voltage and current rating and you just need to know what kind of valve it is. i believe these are 4 way valves.